D Q G 5 H V LQ ' H IH Q V H V LQ 2 OG * UR Z WK 3 R Q G H ... · ( IIH F WV R I 3 UH V F ULE H G % X UQ LQ J R Q 0 R UWD OLW\ D Q G 5 H V LQ ' H IH Q V H V LQ 2 OG * UR Z WK 3 R Q

$Page 1: D Q G 5 H V LQ ' H IH Q V H V LQ 2 OG * UR Z WK 3 R Q G H ... · ( IIH F WV R I 3 UH V F ULE H G % X UQ LQ J R Q 0 R UWD OLW\ D Q G 5 H V LQ ' H IH Q V H V LQ 2 OG * UR Z WK 3 R Q$
Effects of Prescribed Burning on Mortality and ResinDefenses in Old Growth Ponderosa Pine (Crater Lake,Oregon): Four Years of Post-Fire Monitoring

Authors: Perrakis, Daniel D.B., Agee, James K., and Eglitis, Andris

Source: Natural Areas Journal, 31(1) : 14-25

Published By: Natural Areas Association

URL: https://doi.org/10.3375/043.031.0103

BioOne Complete (complete.BioOne.org) is a full-text database of 200 subscribed and open-access titlesin the biological, ecological, and environmental sciences published by nonprofit societies, associations,museums, institutions, and presses.

Your use of this PDF, the BioOne Complete website, and all posted and associated content indicates youracceptance of BioOne’s Terms of Use, available at www.bioone.org/terms-of-use.

Usage of BioOne Complete content is strictly limited to personal, educational, and non - commercial use.Commercial inquiries or rights and permissions requests should be directed to the individual publisher ascopyright holder.

BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofitpublishers, academic institutions, research libraries, and research funders in the common goal of maximizing access tocritical research.

Downloaded From: https://bioone.org/journals/Natural-Areas-Journal on 02 Nov 2020Terms of Use: https://bioone.org/terms-of-use

14 Natural Areas Journal Volume 31 (1), 2011

Natural Areas Journal 31:14–25

•Effects of Prescribed Burning on Mortality and Resin Defenses

in Old Growth Ponderosa Pine

(Crater Lake, Oregon): Four

Years of Post-Fire Monitoring

Daniel D.B. Perrakis1,4

1Fire EcologistWestern & Northern Service Centre

Parks Canada Agency635 8th Avenue SW, Suite 1550

Calgary, Alberta, T2P 3M3, Canada

James K. Agee2

Andris Eglitis3

2 Prof. EmeritusSchool of Forest ResourcesUniversity of Washington

Box 352100Seattle, Washington 98195

3Entomologist, Deschutes National ForestUSDA Forest Service

1230 NE 3rd Street, Suite A-262Bend, Oregon 97701

•

R E S E A R C H A R T I C L E

4 Corresponding author:[email protected]

ABSTRACT: Forests dominated by old growth ponderosa pine (Pinus ponderosa Dougl.) at Crater Lake, Oregon, have been viewed as good candidates for restoration via prescribed burning. Previous burn experiments in this ecosystem observed that ponderosa pine typically survived burning treatments but suffered high post-fire mortality from bark beetle attacks. This paper describes the results of four years of post-fire monitoring of ponderosa pine mortality and resin flow in areas subjected to low intensity spring burning (SB), moderate intensity fall burning (FB), or no burning (unburned controls, UC). Crown vigor estimates, correlated with ring width indices, were also included as a factor in mortality and resin flow analyses. Burn treatment was significant in both ponderosa pine mortality and resin flow, as follows: FB > SB > UC. These results suggest that resin defenses overall did not protect trees from post-fire beetle attacks. Crown vigor was positively related to both survival and resin flow. The relationships between burning, tree vigor, and resin defenses in this study suggest a complex web of interactions. Although some physiological mechanisms are still unconfirmed, these findings suggest that beetles may have been attracted to ponderosa pine following burning, perhaps via the release of volatile resin compounds. Following attraction, resin defenses appear to have been important for protecting trees from beetle attacks, with greater defenses in trees with higher crown vigor and higher growth rates. Management recommendations, including a gradual and incremental approach to fire restoration in these stands, are suggested.

Index terms: bark beetles, mixed-conifer, oleoresin, prescribed burning, vigor

INTRODUCTION

Ponderosa pine (Pinus ponderosa Dougl.) forests of the western and southwestern United States present the classic case for fire restoration. Following over one hun-dred years of fire exclusion, many stands have shown increases in tree density, fuel loading and wildfire severity, and overstory mortality compared to historic conditions (Parsons and DeBenedetti 1979; Agee 1993; Covington and Moore 1994; Sugi-hara et al. 2006). Consequently, widespread interest exists in restoration treatments that mimic elements of the historic disturbance regime while maintaining the presence of dominant individual pines (Thomas and Agee 1986; Sackett and Haase 1998; Friederici 2003; Kolb et al. 2007). Such treatments, including prescribed burning and mechanical fuel reduction, have shown considerable promise in the short term for reducing post-treatment wildfire intensity or severity (Pollet and Omi 2002; Finney et al. 2005; Raymond and Peterson 2005; Stephens et al. 2009). However, long-term success with this strategy remains uncer-tain, and successful restoration may involve more than simply lighting prescribed fires, for example, in fire-excluded areas.

Restoring fire in the mixed-conifer forests of Crater Lake National Park, Oregon, has been a mixed success story so far. In these stands, a fire history study suggested that a low severity fire regime (Agee 1993) ex-

isted during pre-settlement times (McNeil and Zobel 1980). The regime apparently ended following park establishment in 1902, when fire exclusion and land-use changes suddenly and profoundly reduced the influence of fire on park lands. Pre-scribed burning began in the late 1970s and was intended to reduce tree density (particularly of post-settlement firs, Abiesspp.), reduce fuel loading, and maintain larger ponderosa and sugar pine (Pinus lambertiana Dougl.) individuals. After the first decade of burning, however, problems with these treatments became apparent. Specifically, canopy-dominant ponderosa and sugar pines were dying in the first few years following these fires, with most post-fire mortality ultimately attributed to bark beetle attacks, particularly from the western pine beetle (Dendroctonus brevi-comis LeConte, attacking ponderosa pine) and mountain pine beetle (D. ponderosaeHopkins, attacking sugar pine) (Thomas and Agee 1986; Swezy and Agee 1991). Further research at Crater Lake and else-where confirmed that this response pattern is widespread: beetle attacks and beetle-caused tree mortality in ponderosa pine forests increase significantly following prescribed burns or wildfires (Miller and Keen 1960; Ganz et al. 2003; McHugh et al. 2003; Breece et al. 2008). As fire restoration in these forests is frequently intended to maintain large pine individuals, high post-fire tree mortality has become a critical management issue in many areas


Volume 31 (1), 2011 Natural Areas Journal 15

(Kolb et al. 2007; Crater Lake NP staff,pers. comm. 2007).

Much speculation has ensued regardingthe mechanism behind increased post-firebeetle attacks. Although many possibleinteractions could be conceived betweenfire and subsequent beetle attacks in in-jured trees, most discussion has focusedon conifer resin defenses and bark beetlehost selection. (For a review of bark beetlehost selection and ecology, particularlyconcerning D. brevicomis, see Miller andKeen (1960), Wood (1972, 1982), Moeck etal. (1981), Raffa et al. (1993); conifer resindefenses, as well as interactions betweenbeetles and resin, have been reviewed byByers (1995), Phillips and Croteau (1999),Smith (2000), Trapp and Croteau (2001),and Seybold et al. (2000, 2006).

Based on the accumulated body of knowl-edge on these processes, there appear tobe two main possibilities for why westernpine beetles might be more successful atkilling ponderosa pines following fire:(1) reduced tree defenses (and unchangednumber or effectiveness of beetle attacks)or (2) unchanged tree defenses and a greaternumber of total beetle attacks. The latterpossibility implies that fire induces primaryattraction to injured trees (Moeck et al.1981; Wood 1982) or somehow increasesthe potency of the pheromone mixtureemitted by beetles. It is also possible thatboth compromised defenses and increasedattraction occur together.

A recent article on seasonal burning atCrater Lake provided some preliminaryresults from this project (Perrakis andAgee2006). For two years after spring and fallburns in Crater Lake mixed-conifer stands,higher mortality of large ponderosa pineswas observed in hot fall burns than incool spring burns or controls; high vigortrees were also significantly more likelyto survive than low vigor trees, based onvisual assessments of crown vigor. Barkbeetle attacks were associated with mostlarge pine mortality, with western pinebeetle and possibly red turpentine beetle(D. valens LeConte) attacks being the mostimportant species. Part of that study alsoconcerned fire effects on ponderosa pineresin defenses, seeking to identify whether

fire might reduce resin flow or pressure.No evidence of such a mechanism wasdetected, with many burned trees showinghigher resin flow than unburned controlsup to two years after treatment. Thesefindings suggested some important gapsin the state of knowledge on fire-beetleinteractions in this system.

The objective of this study was to continuemonitoring resin defenses and mortalityin this system for a total of four yearsfollowing prescribed burning treatments.Specifically, the study was designed toanswer the following questions: for howlong (up to four years) is the elevatedmortality signal detectable after spring andfall burning treatments in this ecosystem,and how does it relate to tree vigor indicesand different burning treatments? Whatis the nature of the post-fire resin flowresponse in this ecosystem? Does resinflow remain elevated for more than oneyear after burning, return to the level ofunburned control trees, or become reducedcompared to controls, within four yearsafter burning treatment? Answering thesequestions would help provide managerswith specific guidance on restorationtreatments and policies to conserve theselegacy trees in the landscape.

METHODS

Study area and burn treatments

Crater Lake National Park is located at thecrest of the southern Oregon Cascades.The experimental site consisted of an areaof previously unmanaged (except for fireexclusion since 1902) mixed-conifer for-est dominated by old-growth ponderosapines, with a variable-density undergrowthcomposed of several other conifers: whitefir (Abies concolor (Gord. and Glend.)Lindl. ex Hindebr.), Shasta red fir (Abies magnifica var. shastensis Lemm.), lodge-pole pine (Pinus contorta var. murrayana(Greg. and Balf.) Engelm.), and severalother species. The mean diameter at breastheight (dbh) of the ponderosa pine popula-tion was about 90 cm (Perrakis and Agee2006), with very few trees smaller than40 cm dbh; increment cores taken fromseveral samples suggest that the larger trees

were 300 to 400 years old (D. Perrakis,unpubl. data). The 67 – ha study area wasdivided into 24 separate experimental unitsof approximately 2 – 4 ha each, separatedrandomly into spring burn (8), fall burn(8), and control (8) treatments.

Prescribed fire treatments were conductedin spring (late June) and fall (early Oc-tober) 2002. As described previously(Perrakis 2004; Perrakis and Agee 2006),fire behavior and areal coverage in springburns were somewhat less intense thanintended (fire coverage within burn units19% – 57%), while fall burns were closerto desired prescription values (coveragebetween 64% – 86%). Spring burn effectswere confined to the upper forest floor andunderstory, with some lower bole char-ring observed and possible root mortalityexperienced by most overstory pines, butvery little crown scorch or duff consump-tion. In contrast, fall burn effects includedmuch greater consumption of forest floorlayers, extensive bole charring, and oc-casional scorching of the lower crownsof dominant pines (Perrakis 2004). Thus,season of burn was confounded with fireintensity, and the three treatments in thisexperiment can, therefore, be described ascool spring burns (SB), hot fall burns (FB),and unburned controls (UC).

Vigor and mortality monitoring

The ponderosa pine population inside thestudy area included 1725 individuals. Be-fore treatment, we assessed the crown vigorof these trees using a simple categoricalassessment based on the ‘California PineRisk-Rating System’, as described byF.P. Keen (1943). This method involves aqualitative estimate of overall tree vigorbased on crown appearance and approxi-mate age-class, with four classes –A (mostvigorous) through D (least vigorous). Theentire sample of 1725 trees was monitoredannually for mortality during the 2002– 2006 period, and signs of bark beetleattacks, including pitch tubes, frass, gallerypatterns, and feeding by woodpeckers werenoted to identify beetle-related mortality.Although tree diameter data was collected,it was non-significant in previous mortalityanalyses (Perrakis 2004) and so was not



included here; this population consisted en-tirely of mature and old-growth individuals.Likewise, 96 of the trees were subjectedto resin flow sampling (see below) andprevious resin exudation pressure measure-ments. These efforts were believed to havehad negligible effects on the trees and werenot factored into the mortality analysis. Inaddition, half of the resin flow trees werealso treated with a light basal litter rakingtreatment. Although fuel raking may be auseful technique to reduce the heat affect-ing a tree’s lower bole (Hood 2010), thetreatments applied in this study appearedineffective for this purpose, as considerablesmoldering still occurred in the remainingunraked duff (Perrakis 2004). Therefore,the light raking treatments were also omit-ted from analysis in this study.

Increment cores of the outer xylem ringswere collected in summer 2005 from 10trees per experimental unit. In each unit,cores were taken at breast height (1.4 m)from the north sides of five A- or B-classtrees (referring to Keen’s crown vigorclasses), and from five C- or D-class trees;trees were randomly selected within eachunit out of the pool of available A/B andC/D class trees. This sampling strategyresulted in 239 viable cores being col-lected in total (one core was lost duringanalysis). On all cores, the number of ringsin the outer centimeter was counted andMahoney’s Periodic Growth Ratio (PGR)was calculated. The PGR evaluates the ratioof the most recent five years of growth(ring widths) to those of the previous fiveyears, thereby indicating whether radialgrowth has been increasing, stable, ordecreasing within the past decade (Raffaand Berryman 1982).

Resin flow monitoring

Resin flow was measured on a subsampleof the full study population: 96 large pon-derosa pines were monitored on severaloccasions before and after prescribed firetreatments (Perrakis 2004). This group wassplit into 48 each of high vigor and lowvigor trees, with four of these trees (twohigh vigor, two low vigor) located in eachof the 24 experimental units describedabove. Trees were selected randomly

for resin sampling based on crown vigorclasses; the high vigor group consisted ofA-class trees (or B-class, when no A-classtrees existed) and the low vigor groupconsisted of C-class trees.

Resin sampling involved drilling angledholes into each tree bole at approximatelybreast height (1.2 – 1.5 m), with brass fun-nels directing resin into centrifuge tubes,and resin volumes measured 24 hoursafter drilling. Two holes were drilled ineach tree at each measurement time, onapproximately opposite sides of the bole.Pre-treatment resin flow data were col-lected only from the fall burn experimentalunits and half of control treatment units.As a result, the full dataset of resin flowvalues are compared only in terms of totalresin volume at each time period; a separateanalysis based on differences from pre-burnvalues was done only for the fall burn units(see Data Analysis below). Pre-treatmentsampling on fall treatments (burns andfour control units) was conducted in earlySeptember 2002. Measurements in 2003– 2006 were conducted in mid-August ofeach year for consistency, with all mea-surements conducted within a few daysof stable weather to avoid bias. Although2002 – 2004 data were previously presented(Perrakis and Agee 2006), results from allyears are included for completeness.

Data Analysis

The mortality analysis used logistic re-gression on individual tree data (Dalgaard2002). Treatment (SB, FB, UC) and Keen’scrown class (class A, B, C, or D) werethe categorical predictor variables. Themodel was fitted twice: once for the entiresample (N=1725) and once excluding treesthat were obviously killed directly by fire(i.e., trees that burned through at the base),windthrow, or other causes (N=1701). Thesecond run was designed to isolate themajor factor of interest to this study, thatbeing post-fire attacks by bark beetles andother insects or pathogens.

Increment core data were analyzed usinga simple one-way analysis of variance.Mahoney’s PGR and the number of ringsper outer cm were both evaluated separately

as a function of crown vigor class. Post-hoc pairwise differences were evaluatedusing Tukey’s HSD test (Zar 1999) at the

=0.05 level of significance.

The resin flow monitoring data were ana-lyzed using split-plot design linear models(Oehlert 2000; Pinheiro and Bates 2000).Replication (the burn treatments) and treegrouping were at the level of experimen-tal unit, with resin flow measured in thefour sample trees per unit. Resin volumeswere square-root transformed to stabilizevariances before analysis. Categoricalpredictors included Crown class (high orlow, based on Keen’s classes A/B or C),treatment (UC, SB, FB), and year (2003– 2006). Fall burn treatment units and fourcontrol units were also analyzed separatelyas differences from pre-treatment resin flowvalues. For this analysis, pre-treatmentvalues were subtracted in each tree fromeach year’s measured values (post – pre);differences could, therefore, be either posi-tive or negative. Fixed factors (treatment,crown class, and year) and all interactionswere initially included in the model; termswere included in the final model based onthe =0.05 level of significance.

All analyses described in this study wereperformed using R version 2.7 (R – ProjectCollective; available online: http://www.r-project.org), with differences evaluatedat the =0.05 level of significance.

RESULTS

Ponderosa pine mortality

By fall 2006, 139 large ponderosa pineindividuals had died, representing 8.1% ofthe original population of 1725. Of thosethat died, 24 were estimated to have beenkilled directly by burning or windthrowin 2002 or early 2003. Mortality for theremaining 115 trees out of 1701 (6.8%)was at least partly related to post-treatmentinsect or pathogen attack. The majority ofkilled trees occurred in burned treatmentunits (SB or FB; Table 1).

The main effects logistic regression modelwas significant at the =0.05 level whenrun with 1- all trees and 2-only live trees



and those killed by secondary mortality(Table 2). The null model, indicated by theIntercept in Table 2, represents an A-classtree and a control treatment. The treatmentfactor was significant, with spring burningincreasing the probability of mortality andfall burning increasing it further yet. Thecrown class factor was also significant, withclass B, C, and D class trees more likelythan class A trees to die during the study,and coefficient estimates (correspondingpositively to mortality probabilities) in-creasing from class A through D. Despitethe statistical significance of the indepen-dent variables, it should be noted that thepredictive power of the logistic regression

was low, explaining only 10% – 12 % ofobserved mortality based on residual devi-ance (Table 2).

The overall mortality pattern is evidentin Figure 1: probability of mortality wasinversely proportional to crown vigor(as indicated by the crown classes), andincreased by treatment from UC (low-est probability) to SB and FB (highestprobability). The apparent exception tothis is the higher than expected mortalityrate among A-class trees in SB units; thisis probably due to a small sample size– only one tree died in that group out of19, representing 5.3% of the initial group

size. The pattern of increased mortality ascrown vigor decreases is evident from theremaining crown classes in that treatmentgroup, classes B through D.

Figure 2 summarizes the total proportionsof trees killed by year and treatment type.Mortality in spring and fall burn units wasclearly highest within the first year afterburning, and then dropped off steadilyin the following years. By the end of thestudy, 2.3%, 6.1%, and 16.4% of trees incontrol, spring burn, and fall burn treat-ments, respectively, had died.

Tree vigor

The number of cores sampled from thefour crown classes was 16, 97, 110, and16 from classes A through D, respectively– proportions very similar to the study areapopulation as a whole. Mahoney’s PGRwas not significantly related to Keen’scrown classes (p = 0.471). However, therelationship between number of rings perouter centimeter and Keen’s crown classwas highly significant (p < 0.001), withincreasing number of rings/cm betweenclasses A to D (along with high variabilitywithin groups; Figure 3). As the analysissummary and pairwise contrasts show(Table 3), the A – B class and C – D classring width differences were not significant,but all other class contrasts were significant

Table 1. Number (percentage of group total) of dead overstory ponderosa pines by treatment andKeen’s crown vigor class, 2003-2006, excluding direct mortality from fire and windthrow. UC, SBand FB refer to Unburned Control, Spring Burn and Fall Burn treatments, respectively.

Table 2: Parameters of the logistic regression models. The model form is = exp( 0 + 1x1 + 2x2 + ... + qxq), where p is the probability of mortal-

ity, x1 through xq are predictor variables, 0 is the intercept, and 1 through q are the coefficient estimates. See text for factor descriptions. Factorsmarked with an asterisk (*) are significant at the = 0.05 level.

1pp



at the =0.05 level. Mean numbers of ringsper outer cm (and standard deviations)for the four classes were A: 21.8 (11.9);B: 26.4 (13.2); C: 35.8 (15.3); D: 42.3(17.4). The grand mean from all sampleswas 31.5 rings/cm.

Resin flow

Prior to burning, resin flow was notsignificantly different between trees infall burn and control units (Perrakis andAgee 2006). In the first post-treatmentyear (2003), there were several instances

when sampling test tubes overflowed withresin in the highest flow samples. This wasnoted in seven trees in fall burn treatments(four high, three low vigor), four trees inspring burn treatments (three high, onelow vigor), and one tree in control treat-ments (high vigor). In following years, aswitch to larger centrifuge tubes (50 mL)eliminated this problem. Thus, resin flowwas slightly higher in 2003 than the valuespresented, particularly in fall and springburn treatments.

Post-treatment analysis results are pre-sented in Tables 4 and 5. Trt and Cclassrefer to treatment type (Control, SB, or FB)and Keen’s crown class (high vigor or lowvigor) respectively. The intercept representsthe default case of a high vigor tree (Aclass) in a control (unburned) treatmentunit. Over the course of the experiment,two of the 96 trees died – both low vigortrees, one in a spring burn unit in 2003,one in a fall burn unit in 2004. Both treeswere killed by post-fire western pine beetleattacks.

None of the 2-way or 3-way interactionterms was significant, so these weredropped from subsequent model runs.The non-significance (p = 0.84) of the Trt:Year term in a full-factorial run (resultsnot shown) suggests that mean treatmenteffects were not significantly differentbetween the years considered in the study.All fixed-effect factors were significant atthe 0.05 level (Table 4). Both the SB andFB terms were significantly different fromzero, suggesting that trees in both springburn and fall burn treatments had signifi-cantly higher resin flow than in controls.Low vigor (ClassC) trees had significantlylower resin flow than high vigor trees. Theoverall Year effect is somewhat harder tointerpret. Although the fixed factor termwas significant, none of the coefficients forindividual years was different from zeroat the 0.05 level. Year-to-year variationin resin flow likely affected all trees, butthese results suggest that they were not asignificant factor in any individual yearduring this experiment.The separate analysis of resin differencesbetween fall treatments shows similar find-ings (Table 5). Trees in fall burn units hadgreater resin flow differences than those in

Figure 1. Post-treatment ponderosa pine mortality, 2003-2006, grouped by Keen’s (1943) class (A throughD refer to crown vigor classes; see text) and treatment type (Control: unburned; SB: spring burn treat-ment; FB: fall burn treatment).

Figure 2. Percentage of large ponderosa pines killed by year and treatment from all mortality sources(proportion of initial live population).



control units, further suggesting a post-fireincrease in resin flow. In this analysis, theCclass term was not significant, suggesting

that the trees’resin response to burning wassimilar between the high and low crownvigor classes. The Year factor was also not

significant in this analysis, indicating thatrelative resin flow change was constantacross the study timeline.

The overall resin flow trend over the fiveyears of the study is presented in Figure4 (presented as means of all trees in eachtreatment group). The post-treatment re-sponse of increased resin flow in burnedtrees is apparent from the figure, with bothspring and fall burn treatments apparentlypeaking in the second year after burning(2004). By 2006, resin flow from springburn and fall burn treatment groups haddeclined to levels nearly approachingcontrol treatments.

DISCUSSION

Mortality and vigor

This study demonstrated that post-fire mor-tality in southern Cascade ponderosa pineforests could continue for several years.By the end of the monitoring period (fouryears after burning), mortality had declinedin spring burn units to match the level as-sociated with control units (two trees diedin 2006 in each treatment type, 0.3% ofpre-treatment populations). Mortality infall units was declining steadily as well,although was still elevated compared tocontrols (13 trees died in 2006, 2.4% ofpre-treatment population). Continuing thepattern found in previous surveys at thissite (Perrakis and Agee 2006), mortal-ity was strongly associated with Keen’s(1943) crown class. Although this type ofrisk-rating system is coarse and somewhatsubjective, it was significantly (inversely)correlated with the number of rings inthe outer centimeter in these trees. Thus,trees in A and B classes had higher radialgrowth in recent years than those in C orD classes, and the former were more likelyto have survived for four years after burn-ing, regardless of burn season or controltreatment. The other growth index assessedin the model, the PGR, was not relatedto Keen’s crown vigor class. The treesstudied in this experiment were very old(most estimated at > 200 years old) andhad low growth rates. With no exogenousdisturbances and little in-stand changeduring the decade prior to the burning treat-

Table 3. Anova table and Tukey test summary for the increment core analysis (Rings per outer cmas a function of Keen’s crown vigor class). ‘Contrast’ refers to the crown classes being compared,while ‘CI’ represents Confidence Interval. Factors or contrasts marked with an asterisk (*) aresignificant at the = 0.05 level.

Figure 3. Boxplots representing number of rings per outer centimeter, grouped by Keen’s (1943) crownclass. Center lines and boxes represent median and interquartile range values, respectively. Circlesrepresent outlier values in each crown class. Numbers of tree cores analyzed in each class were 16, 97,110, and 16, for classes A through D, respectively.



ments, radial growth of most trees was in along-term pattern of decline, and the PGRvaried only slightly, and inconsistently,between treatment groups.

An interpretation of these two findingssuggests that overall recent radial growthrate may have been indicative of thetrees’ ability to withstand the stresses offire injury and possibly post-fire beetleattacks, while minor variations in growthrate during the past decade did not influ-ence this response pattern. With respect togrowth and vigor, it is unsurprising thata higher vigor index should correspondwith higher survivorship, as trees with

well developed root and crown systemstend to have high carbohydrate reservesand are more resilient to environmentalstresses and pathogen attacks (Waring1987; Kozlowski and Pallardy 1997). Thepresent findings are consistent with thisconceptual model, with high crown vigorclass being positively correlated with widergrowth rings, higher resin flow (see below),and lower mortality rates. Several previousstudies in ponderosa (Larsson et al. 1983)and lodgepole pine (Stuart 1984) have alsofound higher radial growth indices to bepositively related to tree survival followingbark beetle attacks.

The mortality levels observed in this study(2.3%, 6.1%, and 16.4% in control, springburn, and fall burn treatments, respectively,after four years) are similar to or slightlylower than levels reported in other stud-ies on ponderosa pine survival after fire.Thies et al. (2005) studied mortality inuneven-aged stands (7.5 to >75 cm dbh)in eastern Oregon. They noted between 0– 0.6% annual mortality in control plots,and mortality in burn plots that peaked oneyear after burning (spring burns: 4.3%; fallburns: 11.8%) and leveled off (matchingcontrol levels) after four years of post-burnmonitoring – a nearly identical pattern tothat observed in the present study. Their

Table 5. Data summary from resin difference analysis, fall burns and controls. Pre-treatment (early September 2002) resin flow values were subtractedfrom measurements in subsequent years for each tree. Abbreviations are as in Table 4, above. Burn treatments were conducted on 8 experimental units;these are contrasted with 4 control units (4 sample trees per unit). Factors or terms marked with an asterisk (*) are significant at the = 0.05 level.

Table 4. Anova table and summary from post-treatment resin flow measurements, August 2003-2006. DF represents degrees of freedom. The Value columnshows estimates of the linear model coefficients (based on transformed data); Intercept represents Control treatments and A-class (high vigor) trees; SB(spring burn), FB (fall burn), and ClassC model terms and statistics are contrasted with high vigor (Keen’s class A) trees in control units. Factors or termsmarked with an asterisk (*) are significant at the = 0.05 level.



fall burn treatments caused higher mortal-ity overall (29% of trees dead after fouryears), but the uneven-aged populationstructure suggests that many of the killedtrees may have been much smaller thanthose in the present study and, therefore,susceptible to mortality from crown scorch(Harrington 1993).

As reported previously, earlier studies atCrater Lake observed higher mortality inspring burns than in fall burns (Swezy andAgee 1991;Agee 2003). However, a reviewof the literature on ponderosa pine mortalityreveals mixed findings on seasonal fire ef-fects.Although phenology may suggest thatearly season burn effects are most likelyto be lethal due to reduced carbohydratestores (Kozlowski 1992), prescribed burnslater in the season tend to have higherfire intensities and proportional coverage(Ganz et al. 2003; Knapp and Keeley2006; Perrakis and Agee 2006; Schwilket al. 2006). A recent study comparingspring and fall burning in a Sierra Nevadamixed-conifer forest found that ponderosaand sugar pine mortality was significantlyrelated to fire intensity, while mortality was

not significantly different overall betweenthe two burning seasons (Schwilk et al.2006). In northern Arizona, McHugh andKolb (2003) compared spring and earlysummer wildfires to a fall prescribed fire.They observed that the most significantpredictors of mortality in logistic regres-sion models were crown damage (scorchand consumption) and bole char – factorsrelated to fire intensity rather than burnseason. In the present study, with springburns having significantly lower coverageand visibly lower intensities than fall burns,a similar pattern emerged.Although seasonof burn and tree phenology should not bediscounted, the intensity of a particularfire is probably a more important variableoverall in predicting mortality.

For management purposes, it is promis-ing that mortality has declined steadilyeach year after post-burn year 1, even inthe intense fall burn treatments. Thirteenyears after a series of very similar burnsat another Crater Lake site, Agee (2003)found no significant difference betweenlarge ponderosa pine mortality in burnedunits and unburned controls. Ideally, post-

burn mortality should be below controlunit mortality levels after a few years, inkeeping with the intent of the treatments.As previously discussed, these sites areconsidered candidates for restorationdue to the large increases in tree densityand competition in the fire-exclusion era(McNeil and Zobel 1980; Agee 2003);accordingly, burning treatment objectivestend to include increased survivorshipamong remaining old pines (Kolb et al.2007). Reaching a point of equal mortal-ity levels between burn treatments andcontrols after a few years is a good start,but is not indicative of long-term successaccording to these criteria. Regardless ofmortality levels, ponderosa pines in thesestands are old, with very low survival andrecruitment of new seedlings in unburnedareas (Perrakis and Agee 2006; D. Per-rakis unpubl. vegetation data). Ponderosapine is well-known to require open standconditions and mineral soils to regeneratenaturally from seed (Habeck 1992), andthese combined factors emphasize the needfor restoration treatments in these stands,despite risks to the overstory trees.

Resin flow monitoring

This study presents one of the longestannually repeated post-fire resin monitor-ing efforts in the literature. The pattern ofincreased post-fire resin flow identified inseveral other studies (Santoro et al. 2001;Lombardero et al. 2006; Knebel andWentworth 2007) is hereby confirmed tocontinue for four years (or more) after firein old-growth ponderosa pines. The resultssuggest that four years after burning, withsome fire injury to tree boles and possiblyroot crowns (but very low crown scorch),bole resin flow remains higher in burnedtrees than those in unburned controls. Therelative increase due to fire injury appearsto be declining in the spring burns butremains elevated among trees in fall burntreatments. Although the data suggest thatresin flow peaked in year two after burning(2004) for both spring and fall burn treat-ments, between-year differences were notstatistically significant and some problemswith overflowing test tubes in year 1 mayhave masked higher resin flow values in thatyear. The analysis of differences between

Figure 4. Resin flow means (untransformed) and standard errors (of unit means) by year and treatmentgroup. Dashed vertical line indicates approximate timing of burn treatments (not to scale); Control:unburned control treatments; SB: spring burn treatments; FB: fall burn treatments.*Note: unbiased data from spring burn treatment trees in 2002 (pre- and post-treatment) was not col-lected and is not presented. Control treatment mean in 2002 is a combination of early August and earlySeptember measurements; 2002 fall burn (pre-treatment) data was collected in early September. Measure-ments in all subsequent years were taken approximately synchronously between treatment groups.



post- and pre-treatment resin flow amongfall burns and controls followed the samepattern of a greater resin increase amongburned trees than controls.

Crown vigor was also a significant factorin the model, with greater absolute resinflow in trees of higher crown vigor class.In contrast, vigor was not a significantpredictor of the post-treatment change inresin flow (among fall-burned trees andcontrols). As crown vigor class was in-versely correlated with rings per cm, thissuggests a positive correlation betweenconstitutive resin flow and sapwood xylemproduction. This finding is in agreementwith the work of McDowell et al. (2007),who noted positive correlations betweenbasal area increment (BAI) and resin flowin a survey of several different experimentsin southwestern ponderosa pine stands.

There has been considerable interest inrecent decades in using plant defensetheories to explain relationships betweenconifer trees and bark beetles. In particular,the Growth-Differentiation Balance (GDB)hypothesis (Loomis 1953; Herms andMattson 1992) has been used in a varietyof studies on Pinaceae trees to explain(or attempt to) carbohydrate allocationtradeoffs between growth processes andinvestment in defenses (Lorio 1986; Lorioet al. 1990; Johnson et al. 1997; Gaylord etal. 2007). According to the GDB, conifersprioritize investment in growth processes(cell division and expansion) over invest-ments in defenses during typical favorablegrowing conditions. The theoretical pattern,therefore, results in an inverse relationshipbetween growth and defense investments,and conversely suggests that plants maybuild stronger defenses during periods ofsub-optimal growth. Although many stud-ies appear to confirm that GDB principlesare sound (see Herms and Mattson 1992;Stamp 2003), studies on conifer resin havebeen equivocal with respect to GDB predic-tions. This study supports a more complexinterpretation than is suggested from theGDB, as increased resin flow did not ap-pear to be associated with resource scarcityor sub-optimal growth. As McDowell etal. (2007) previously observed, increasedresin flow in ponderosa pine appears to beassociated with wider xylem rings, likely

because sapwood is rich in resin canals.The interconnected architecture and long-lived nature of resin canals in pine phloemand xylem tissues (Fahn 1979; Lapashaand Wheeler 1990; Lin et al. 2002) alsosuggest that physiological inferences aredifficult to assess from treatment effects:short-term resin measurements will notnecessarily be associated with the condi-tions during which the resin canals (orresin itself) were produced (Gaylord et al.2007; Wallin et al. 2008).

There remains the issue of what is causingthe apparent increase in resin flow amongburned trees. Little is known in this spe-cies about the sensitivity of resin canaldevelopment and oleoresin production tovariable growing conditions or injury. InPinus taeda, there is some evidence thatpre-formed canals may not be filled tocapacity during normal growing condi-tions. Following physical wounding to thebole, canals are rapidly filled with inducedresin from surrounding epithelial cells,with production most vigorous in the fast-est-growing trees with the largest crowns(Blanche et al. 1992; Lombardero et al.2000). A similar mechanism may occur inP. resinosa as well, albeit on a slower timescale (Lombardero et al. 2006). In Cascadeponderosa pine stands, recent experimentscompared the effects of fire and fire-sur-rogate treatments (bole charring, crownpruning, root trenching) on resin flow. Inthese trials, only the prescribed fire and bolecharring treatments caused post-treatmentresin increases, suggesting that bole heatingor injury induce the resin response, ratherthan any changes in resource availability(Perrakis 2008; D. Perrakis and J. Agee,unpubl. data). However, in similar studiesin Arizona, thinning treatments (with noreported injury to study trees) also appearedto cause a short term increase in resin flow(Wallin et al. 2008). The latter study usedconsiderably smaller and younger trees(which are known to respond more rapidlyto resource availability) than the presentexperiment, which may explain the differ-ence (Kolb et al. 2007).

In this study, treatments associated withgreatest resin flow were not those withgreatest overstory ponderosa pine survival.This suggests that the resin defense hypoth-

esis must be reexamined in this ecosystem.Primary attraction (Wood 1982) of westernpine beetles to stressed ponderosa pinesor resin samples has not been confirmed(Moeck et al. 1981). However, one of theconstituent resin volatiles, myrcene, haslong been known to be an integral part ofthe attractant pheromone mixture for thesebeetles (Bedard et al. 1969;Vité and Pitman1969; Gaylord et al. 2006). Many othervolatile compounds have been confirmedto cause antennal response in westernpine beetles (Shepherd et al. 2007). Thus,we may expect that fire, through damageto live trees or volatization of resin inconsumed woody debris, may increase therelease of these volatiles and thereby attractbeetles, particularly if beetle activity (and,therefore, other pheromone precursors) isalready present in the canopy airspace.Such an effect (increased ambient resinvolatiles) has been observed following thin-ning and chipping treatments (Schade andGoldstein 2003; Fettig et al. 2006), and issuggested as an explanation for post-treat-ment increases in beetle activity (Fettig etal. 2006). This suggested mechanism is alsoconsistent with studies that observed posi-tive relationships between scorch severityand number of beetle attacks in ponderosapine forests (McHugh et al. 2003; Wallinet al. 2003).

The defensive role of resin flow cannotcompletely be rejected, however. The groupproducing the greatest volume of post-treat-ment resin (high vigor trees in intense fallburn treatments) were not those sufferinggreatest mortality, as the higher vigor treesoverall survived the treatments much betterthan low vigor individuals (Table 1). Wesuggest that beetles were attracted to thestand via emitted resin volatiles during andafter burning treatments, but that resin flowin individual trees remained important indetermining the mortality and survival ofindividual trees.

CONCLUSIONS AND MANAGEMENT RECOMMENDATIONS

The results from this study have implica-tions for both managers interested in firerestoration as well as researchers studyingthe mechanisms behind fire/bark beetleinteractions. The main management-related



finding from this study was that survivor-ship of old pines was positively related tohigh radial growth/BAI and high crownvigor. Treatments that improve individualtree vigor, such as removal of compet-ing vegetation surrounding old pines, arelikely beneficial to overstory survival.Additionally, resin flow monitoring inponderosa pine is not a reliable indica-tor of the likelihood of survival after fireand is not suggested as a post-treatmentmonitoring technique.

There may also be merit to minimizing therelease of resin volatiles in stand treatmentsor attempting other measures to mitigatebole injury, although these suggestions aresomewhat speculative. Such actions mayinclude avoiding injuring trees of concernthrough ignition patterns or ‘cooling’segments of a fireline, raking fuel fromthe bases of trees prior to burning (Hood2010), disposing of thinned biomass mate-rial off-site, restricting thinning treatmentsto periods of low beetle activity (fall orwinter), and possibly setting conservative(low intensity) burn prescriptions.

Overall, the findings of this study supportthe notion of an incremental approach tofire restoration in ponderosa pine forestsat Crater Lake. Restoration treatments,including burning, and possibly thin-ning or other mechanical fuel manipula-tions, should be implemented gradually.Although such treatments can probablyeventually benefit these stands, it may benecessary to improve the vigor of indi-vidual trees prior to implementing all butthe least intense prescribed fires, lest barkbeetle populations benefit more than thetrees of concern.

ACKNOWLEDGMENTS

We would like to thank the staff at CraterLake NP, particularly Michael Murray andScott Girdner, for assistance and logisticalsupport over the duration of this project.Dallas Anderson, Mike Keim, Phil Mon-santo, Christine Hurst, and Rob Baneshelped with fieldwork, while Phil Mon-santo, Carson Sprenger, and Tania Taipaleassisted with laboratory analyses. The Joint

Fire Science Program, project 05-2-1-92,funded this study with additional supportprovided by the Western & Northern Ser-vice Centre, Parks Canada Agency; we aregrateful for their contributions.

Daniel Perrakis is a fire ecologist with the Western & Northern Service Centre, Parks Canada Agency, Calgary, AB. In addition to fire-bark beetle interactions, his research interests include burn severity monitor-ing using satellite imagery, fire behavior modeling and prediction, and restoration of historic fire regimes.

James K. Agee is Professor Emeritus of forest ecology at the School of Forest Re-sources, University of Washington, Seattle, WA. He has studied fire history and effects in the western United States for the past 40 years and has authored more than 200 publications on these topics.

Andris Eglitis is an entomologist with the USDA Forest Service, situated on the Deschutes National Forest, Bend, OR. He provides technical assistance and training on forest insects to several federal land management agencies in central Oregon.

LITERATURE CITED

Agee, J.K. 1993. Fire Ecology of PacificNorthwest Forests. Island Press, Washing-ton, D.C.

Agee, J.K. 2003. Monitoring post-fire treemortality in mixed-conifer forests of Cra-ter Lake, Oregon. Natural Areas Journal23:114-120.

Bedard, W.D., P.E. Tilden, D.L. Wood, R.M.Silverstein, R.G. Brownlee, and J.O. Rodin.1969. Western pine beetle: field response toits sex pheromone and a synergistic hostterpene, myrcene. Science 164:1284-1285.

Blanche, C.A., P.L.J. Lorio, R.A. Sommers, J.D.Hodges, and T.E. Nebeker. 1992. Seasonalcambial growth and development of loblollypine: xylem formation, inner bark chemistry,resin ducts, and resin flow. Forest Ecologyand Management 49:151-165.

Breece, C.R., T.E. Kolb, B.G. Dickson, J.D. Mc-Millin, and K.M. Clancy. 2008. Prescribedfire effects on bark beetle activity and treemortality in southwestern ponderosa pineforests. Forest Ecology and Management255:119-128.

Byers, J. 1995. Host-tree chemistry affectingcolonization in bark beetles. Pp. 155-213in R.T. Carde and W.J. Bell, eds., ChemicalEcology of Insects 2. Chapman and Hall,New York.

Covington, W.W., and M.M. Moore. 1994.Southwestern ponderosa pine forest struc-ture: changes since Euro-American settle-ment. Journal of Forestry 92:39-47.

Dalgaard, P. 2002. Introductory Statistics WithR. Springer, New York.

Fahn, A. 1979. Secretory Tissues in Plants.Academic Press, New York.

Fettig, C.J., J.D. McMillin, J.A. Anhold, S.M.Hamud, R.R. Borys, C.P. Dabney, and S.J.Seybold. 2006. The effects of mechanicalfuel reduction treatments on the activityof bark beetles (Coleoptera: Scolytidae)infesting ponderosa pine. Forest Ecologyand Management 230:55-68.

Finney, M.A., C.W. McHugh, and I.C. Grenfell.2005. Stand- and landscape-level effects ofprescribed burning on two Arizona wild-fires. Canadian Journal of Forest Research35:1714-1722.

Friederici, P. 2003. Ecological Restoration ofSouthwestern Ponderosa Pine Forests. IslandPress, Washington, D.C.

Ganz, D.J., D.L. Dahlsten, and P.J. Shea. 2003.The post-burning response of bark beetles toprescribed burning treatments. Pp. 143-158in P. Omi and L.A., Joyce, eds., Fire, Fuelsand Ecological Restoration: ConferenceProceedings, 16-18 April 2002. RMRS-P-29, U.S. Department of Agriculture, ForestService Rocky Mountain Research Station,Fort Collins, Colo.

Gaylord, M.L., T.E. Kolb, K.F. Wallin, andM.R. Wagner. 2006. Seasonality and lurepreference of bark beetles (Curculionidae:Scolytinae) and associates in a northern Ari-zona ponderosa pine forest. EnvironmentalEntomology 35:37-47.

Gaylord, M.L., T.E. Kolb, K.F. Wallin, andM.R. Wagner. 2007. Seasonal dynamics oftree growth, physiology, and resin defensesin a northern Arizona ponderosa pine for-est. Canadian Journal of Forest Research37:1173-1183.

Habeck, R.J. 1992. Pinus ponderosa var. pon-derosa. In Fire Effects Information System.U.S. Department of Agriculture, ForestService, Rocky Mountain Research Station,Fire Sciences Laboratory. Available online:<http://www.fs.fed.us/database/feis>. Ac-cessed 1 June 2009.

Harrington, M.G. 1993. Predicting Pinus pon-derosa mortality from dormant season andgrowing season fire injury. InternationalJournal of Wildland Fire 3:65-72.



Herms, D.A., and W.J. Mattson. 1992. Thedilemma of plants: to grow or defend. Quar-terly Review of Biology 67:283-335.

Hood, S. 2010. Mitigating old tree mortalityin long-unburned, fire-dependent forests: asynthesis. General Technical Report RMRS-GTR-238, U.S. Department of Agriculture,Forest Service, Rocky Mountain ResearchStation, Ft. Collins, Colo.

Johnson, R.H., B.L. Young, and D.N. Alstad.1997. Responses of ponderosa pine growthand volatile terpene concentrations to ma-nipulation of soil water and sunlight avail-ability. Canadian Journal of Forest Research27:1794-1804.

Keen, F.P. 1943. Ponderosa pine tree classesredefined. Journal of Forestry 41:249-253.

Knapp, E.E., and J.E. Keeley. 2006. Heterogene-ity in fire severity within early season andlate season prescribed burns in a mixed-coni-fer forest. International Journal of WildlandFire 15:37-45.

Knebel, L., and T.R. Wentworth. 2007. Influ-ence of fire and southern pine beetle onpine-dominated forests in the LinvilleGorge Wilderness, North Carolina. Castanea72:214-225.

Kolb, T.E., J.K. Agee, P.Z. Fulé, N.G. Mc-Dowell, K. Pearson, A. Sala, and R.H.Waring. 2007. Perpetuating old ponderosapine. Forest Ecology and Management249:141-157.

Kozlowski, T.T. 1992. Carbohydrate sourcesand sinks in woody plants. Botanical Review58:107-222.

Kozlowski, T.T., and S.G. Pallardy. 1997.Growth Control in Woody Plants.AcademicPress, New York.

Lapasha, C.A., and E.A. Wheeler. 1990. Resincanals in Pinus taeda: longitudinal canallengths and interconnections between lon-gitudinal and radial canals. InternationalAssociation of Wood Anatomists Bulletin11:227-238.

Larsson, S., R. Oren, R.H. Waring, and J.W.Barrett. 1983. Attacks of mountain pinebeetle as related to tree vigor of ponderosapine. Forest Science 29:395-402.

Lin, J., Y. Hu, X. He, and R. Ceulemans. 2002.Systematic survey of resin canals in Pina-ceae. Belgian Journal of Botany 135:3-14.

Lombardero, M.J., M.P. Ayres, and B.D. Ayres.2006. Effects of fire and mechanical wound-ing on Pinus resinosa resin defenses, beetleattacks, and pathogens. Forest Ecology andManagement 225:349-358.

Lombardero, M.J., M.P.Ayres, P.L.J. Lorio, andJ.J. Ruel. 2000. Environmental effects onconstitutive and inducible resin defenses ofPinus taeda. Ecology Letters 3:329-339.

Loomis, W.E. 1953. Growth and Differentiationin Plants. Iowa State College Press, Ames.

Lorio, P.L.J. 1986. Growth-differentiation bal-ance: a basis for understanding southern pinebeetle-tree interactions. Forest Ecology andManagement 14:259-273.

Lorio, P.L.J., R.A. Sommers, C.A. Blanche,J.D. Hodges, and T.E. Nebeker. 1990.Modeling pine resistance to bark beetlesbased on growth and differentiation balanceprinciples. Pp. 402-409 in R.K. Dixon, G.A.Meldahl, G.A. Ruark, and W.G. Warren,eds., Process Modeling of Forest GrowthResponses to Environmental Stress. TimberPress, Portland, Ore.

McDowell, N.G., H.D. Adams, J.D. Bailey, andT.E. Kolb. 2007. The role of stand densityon growth efficiency, leaf area index, andresin flow in southwestern ponderosa pineforests. Canadian Journal of Forest Research37:343-355.

McHugh, C.W., and T.E. Kolb. 2003. Ponderosapine mortality following fire in northernArizona. International Journal of WildlandFire 12:7-22.

McHugh, C.W., T.E. Kolb, and J.L. Wilson.2003. Bark beetle attacks on ponderosa pinefollowing fire in northern Arizona. Environ-mental Entomology 32:510-522.

McNeil, R.C., and D.B. Zobel. 1980.Vegetationand fire history of a ponderosa pine-whitefir forest in Crater Lake National Park.Northwest Science 54:30-46.

Miller, J.M., and F.P. Keen. 1960. Biology andcontrol of the western pine beetle. U.S. De-partment of Agriculture, Washington, D.C.

Moeck, H.A., D.L. Wood, and K.Q.J. Lindahl.1981. Host selection behavior of barkbeetles (coleoptera: scolytidae) attackingPinus ponderosa, with special emphasison the western pine beetle, Dendroctonus brevicomis. Journal of Chemical Ecology7:49-83.

Oehlert, G.W. 2000. A First Course in Designand Analysis of Experiments. W.H. Free-man, New York.

Parsons, D.J., and S.H. DeBenedetti. 1979. Im-pact of fire-suppression on a mixed-coniferforest. Forest Ecology and Management2:21-33.

Perrakis, D.D.B. 2004. Seasonal fire effectson mixed-conifer forest structure and pineresin properties. M.S. thesis, University ofWashington, Seattle.

Perrakis, D.D.B. 2008. Fire-bark beetle interac-tions: exploring links between fire injury,resin defenses, and beetle-induced mortal-ity in ponderosa pine forests. Ph.D. diss,University of Washington, Seattle.

Perrakis, D.D.B., and J.K. Agee. 2006. Sea-sonal fire effects on mixed-conifer forest

structure and ponderosa pine resin proper-ties. Canadian Journal of Forest Research36:238-254.

Phillips, M.A., and R.B. Croteau. 1999. Resin-based defenses in conifers. Trends in PlantScience 4:184-190.

Pinheiro, J.C., and D.M. Bates. 2000. Mixed-effects Models in S and S-PLUS. Springer-Verlag, New York

Pollet, J., and P.N. Omi. 2002. Effect of thinningand prescribed burning on crown fire sever-ity in ponderosa pine forests. InternationalJournal of Wildland Fire 11:1-10.

Raffa, K.F., and A.A. Berryman. 1982. Physi-ological differences between lodgepole pinesresistant and susceptible to the mountainpine beetle and associated microorganisms.Environmental Entomology 11:486-492.

Raffa, K.F., T.W. Phillips, and S.M. Salom.1993. Strategies and mechanisms of hostcolonization by bark beetles. Pp.103-128in T.D. Schowalter and G.M. Filip, eds.,Beetle-pathogen Interactions in ConiferForests. Academic Press, San Diego.

Raymond, C.L., and D.L. Peterson. 2005.Fuel treatments alter the effects of wild-fire in a mixed-evergreen forest, Oregon,USA. Canadian Journal of Forest Research35:2981-2995.

Sackett, S.S., and S.M. Haase. 1998. Two casehistories for using prescribed fire to restoreponderosa pine ecosystems in northernArizona. Pp. 380-389 in Fire in ecosystemmanagement: shifting the paradigm fromsuppression to prescription. Tall TimbersFire Ecology Conference Proceedings, No.20, Tall Timbers Research Station, Tal-lahassee, Fla.

Santoro, A.E., M.J. Lombardero, M.P. Ayres,and J.J. Ruel. 2001. Interaction betweenfire and bark beetles in an old growth pineforest. Forest Ecology and Management144:245-254.

Schade, G.W., and A.H. Goldstein. 2003.Increase of monoterpene emissions froma pine plantation as a result of mechanicaldisturbances. Geophysical Research Letters30:1380-1383.

Schwilk, D.W., E.E. Knapp, S.M. Ferrenberg,J.E. Keeley, and A.C. Caprio. 2006. Treemortality from fire and bark beetles follow-ing early and late season prescribed fires ina Sierra Nevada mixed-conifer forest. ForestEcology and Management 232:36-45.

Seybold, S.J., J. Bohlmann, and K.F. Raffa.2000. Biosynthesis of coniferophagous barkbeetle pheromones and conifer isoprenoids:evolutionary perspective and synthesis. Ca-nadian Entomologist 132:697-753.

Seybold, S.J., D.P.W. Huber, J.C. Lee, A.D.Graves, and J. Bohlmann. 2006. Pine mono-



terpenes and pine bark beetles: a marriageof convenience for defense and chemicalcommunication. Phytochemistry Reviews5:143-178.

Shepherd, W.P., D.P.W. Huber, S.J. Seybold,and C.J. Fettig. 2007. Antennal responsesof the western pine beetle, Dendroctonus brevicomis (Coleoptera: Curculionidae),to stem volatiles of its primary host, Pinus ponderosa, and nine sympatric nonhostangiosperms and conifers. Chemoecology17:209-221.

Smith, R. H. 2000. Xylem monoterpenes ofpines: distribution, variation, genetics,function. General Technical Report PSW-GTR-177, U.S. Department of Agriculture,Forest Service, Pacific Southwest ResearchStation, Berkeley, Calif.

Stamp, N. 2003. Out of the quagmire of plantdefense hypotheses. Quarterly Review ofBiology 78:23-55.

Stephens, S.L., J.J. Moghaddas, C.B. Edminster,C.E. Fiedler, S.M. Haase, M.G. Harrington,J.E. Keeley, E.E. Knapp, J.D. McIver, K.Metlen, C.N. Skinner, and A. Youngblood.2009. Fire treatment effects on vegetationstructure, fuels, and potential fire severityin western U.S. forests. Ecological Applica-tions 19:305-320.

Stuart, J.D. 1984. Hazard rating of lodgepolepine stands to mountain pine beetle out-breaks in southcentral Oregon. CanadianJournal of Forest Research 14:666-671.

Sugihara, N.G., J.W. van Wagtendonk, K.E.Shaffer, J. Fites-Kaufman, and A.E. Thode(eds.). 2006. Fire in California’s Ecosystems.University of California Press, Berkeley.

Swezy, D.M., and J.K. Agee. 1991. Prescribed-fire effects on fine-root and tree mortalityin old-growth ponderosa pine. CanadianJournal of Forest Research 21:626-634.

Thies, W.G., D.J. Westlind, and M. Loewen.2005. Season of prescribed burn in pon-derosa pine forests in eastern Oregon: impacton pine mortality. International Journal ofWildland Fire 14:223-231.

Thomas, T.L., and J.K. Agee. 1986. Prescribedfire effects on mixed conifer forest structureat Crater Lake, Oregon. Canadian Journalof Forest Research 16:1082-1087.

Trapp, S., and R. Croteau. 2001. Defensive resinbiosynthesis in conifers. Annual Review ofPlant Physiology and Plant Molecular Biol-ogy 52:689-724.

Vité, J.P., and G.B. Pitman. 1969. Insect andhost odors in the aggregation of the west-ern pine beetle. Canadian Entomologist

101:113-117.

Wallin, K.F., T.E. Kolb, K.R. Skov, and M.R.Wagner. 2003. Effects of crown scorch onponderosa pine resistance to bark beetles innorthern Arizona. Environmental Entomol-ogy 32:652-661.

Wallin, K.F., T.E. Kolb, K.R. Skov, and M.R.Wagner. 2008. Forest management treat-ments, tree resistance, and bark beetleresource utilization in ponderosa pine forestsof northern Arizona. Forest Ecology andManagement 255:3263-3269.

Waring, R.H. 1987. Characteristics of trees pre-disposed to die. Bioscience 37:569-574.

Wood, D.L. 1972. Selection and colonization ofponderosa pine by bark beetles. Pp. 101-117in H.F. van Emden, ed. Insect/Plant Relation-ships. Halsted Press, New York.

Wood, D.L. 1982. The role of pheromones,kairomones, and allomones in the hostselection and colonization behavior of barkbeetles. Annual Review of Entomology27:411-446.

Zar, J.H. 1999. Biostatistical Analysis, 4th ed.Prentice Hall, Upper Saddle River, N.J.


Documents

D Q G 5 H V LQ ' H IH Q V H V LQ 2 OG * UR Z WK 3 R Q G H ... · ( IIH F WV R I 3 UH V F ULE H G % X UQ LQ J R Q 0 R UWD OLW\ D Q G 5 H V LQ ' H IH Q V H V LQ 2 OG * UR Z WK 3 R Q